How to Choose the Right CRC for High Confidence Error Detection?

Join the discussion
Ask a follow-up here, or get your own question answered by working scientists, mathematicians and engineers — people, not an autocomplete.
Real named experts · corrections over time · the nuance an AI answer skips
1 reply · 3K views
Emanresu
Messages
53
Reaction score
0
Hi,

I have a small fixed size packet of information that I want to transmit and detect errors with a high degree of confidence (no correction is required). I'm guessing that a CRC is probably the way to go but don't know how to work out what form of CRC to use. I want to be able to specify a value for the likelihood of failure to detect an error.

If for example I wanted the likelihood of failure to detect an error to be 1 in a billion when the probability of an error is high, say 10%, how do I work out how to achieve this ?

E.
 
Physics news on Phys.org
Shannon's theorem http://en.wikipedia.org/wiki/Shannon's_theorem tells you how much error correction you need for a given degree of failure on a noisy channel.
In practice it largely depends on if you expect the noise to be fairly constant, changing bits in each word, or bursty, corrupting long sequences of data.
The wiki link has good links to different error correcting codes.